Background The RAID is a patient reported outcome (PRO) measure evaluating the impact of rheumatoid arthritis (RA) on patients' quality of life. It comprises 7 domains (pain, function, fatigue, physical and psychological wellbeing sleep disturbance and coping), that are evaluated as continuous variables from 0 (best) to 10 (worst) , validated predominantly in Caucasians. The final score ranges from 0-10. It has not been applied in Asian populations, who may have different reporting of PROs and life impact.
Objectives To evaluate the metrological properties of RAID in a multiethnic Asian RA cohort, and to compare these properties to those reported in the initial study of the RAID in a predominantly Caucasian population.
Methods Cross-sectional assessment of consecutive RA patients in a Singapore tertiary centre was performed. Validity was assessed by measuring correlation with patient global assessment (PGA), disease activity score in 28 joints (DAS28) and short form 12 (SF-12) (Spearman's). Consistency and reliability were assessed by Cronbach's alpha and by intra-class correlation coefficient (ICC) respectively. The cut-off for patient acceptable state was evaluated by receiver operator curve (ROC) analysis. To determine the variance in RAID attributable to different factors (calculation of R2), two separate multiple linear regressions were performed, one with patient-related explanatory variables (i.e., ethnicity, education, sex, body mass index or BMI, age, smoking) and the other with disease-related variables (i.e., DAS28, the 8-question modified-Health Assessment Questionnaire, mHAQ, SF-12, PGA, presence of either rheumatoid factor or anti-CCP).
Results In all 101 RA patients (76% Chinese, 12% Malay, 12% Indian) were analysed: mean age 54.5±12.8 yrs, disease duration 6.3±6.0 yrs, and 80% women. Mean RAID score was 2.5±2.0, DAS28 was 3.3±1.1 and mHAQ was 0.1±0.3. RAID was strongly correlated with PGA (R=0.75), and weakly but significantly correlated with other outcomes (DAS28 R=0.44, SF12 physical −0.40 and mental −0.50, p<0.0001 for all). Consistency and reliability was high with Cronbach's alpha=0.91 and ICC=0.84 (95% CI: 0.74–0.90) respectively. The cut-off point for a patient acceptable symptom state was below 2.24. By linear regression, demographic variables explained 2.2% of the variance of the RAID whereas disease characteristics and activity explained 65.9% of the variance of the RAID.
Conclusions The RAID may be a valid and practical PRO to assess the impact of RA in Asia. The metrological properties of RAID in this multiethnic Asian cohort are comparable to that of the original RAID-development Caucasian RA population, although sensitivity to change needs to be further evaluated.
Gossec et al. Ann Rheum Dis 2009;68:1680-5.
Disclosure of Interest None declared